The disproportionate burden of HIV/AIDS among African American men who have sex with men (MSM) is stark and alarming, and the lack of an effective preventive vaccine or microbicide emphasizes the urgent need for effective behavioral interventions. Although considerable research has been conducted to develop behavioral interventions for MSM in general, few interventions have been developed exclusively for African American MSM whose social community involves a complex intersection of economics, ethnicity, family ties, community values, and personal aspirations, all of which impact on their intimate partnerships and sexual behavior. The specific aim of this study is to test the efficacy of a group-level, Social Cognitive Theory-based sexual risk-reduction intervention for African American HIV infected and uninfected MSM. The intervention weaves proven cognitive-behavioral sexual risk-reduction information and skills-building modules into cooking and meal-sharing activities to engage participants in creating supportive caring communities of friends that value health, self-worth, self-identification as MSM, and social integration. This is a novel ecological approach to HIV risk-reduction in that the intervention re-engineers the social environment and models interpersonal socialization that can be replicated by participants within their own social contexts. [unreadable] [unreadable] To accomplish the aim, we will build upon our partnerships with two community-based agencies in New York City and propose to conduct a randomized, controlled trial among 300 sexually active HIV infected and uninfected African American MSM. Study participants will be randomized to the experimental or control condition in a 1:1 ratio. At baseline and 3 months, standardized data collection using audio computer assisted self-interviewing (ACASI) technology will assess outcomes and covariates. The primary outcome will be the occurrence of unprotected receptive anal sex (URA) and unprotected insertive anal sex (UIA). Secondary outcomes will include 1) number of new and total male sex partners; 2) occurrence of URA and UIA while under the influence of substances; and 3) level of identity integration and self-efficacy for safer sex. This study also will assess moderators of the effect of the intervention on behavior change and other factors identified as important influences of risk behaviors will be assessed. The hypothesis to be tested is that a culturally appropriate, theory-based behavioral intervention that addresses sexual risk and related factors will reduce sexual risk behaviors among African American MSM. [unreadable] [unreadable] [unreadable]